 |
 |
I would think that you are better off just using flirt (without the mcflirt step for the same diff directions). I think this because if the head moves in the image, the effect of eddy currents on the "brain space" will change, making a rigid body transformation insufficient to match the two brains. Shouldn't think it will make that much difference though. Cheers T ------------------------------------------------------------------------------- Tim Behrens Centre for Functional MRI of the Brain The John Radcliffe Hospital Headley Way Oxford OX3 9DU Oxford University Work 01865 222782 Mobile 07980 884537 ------------------------------------------------------------------------------- On Sun, 11 Jul 2004, Stephen Smith wrote: > Hi - yes, that does sound sensible. These aren't things we've investigated > in great detail. I guess if you have a lot of head motion then you might > need affine as well for within-direction averaging, if there was > interaction between eddies and head motion. Another finesse is that your > approach will end up interpolating each image twice, and with a bit more > bookkeeping you should be able to avoid that - not sure if it's worth the > effort though. Any thoughts on that, MJ or Tim? > > Cheers. > > > > > On Thu, 8 Jul 2004, Russ Poldrack wrote: > > > > > > > Strange - this must be a platform-dependent bug, as it works fine on my > > > mactop, and gives good results too. What platform are you using? > > > > > > However, anyway: the attached script eddy_correct (which will be part > > > of > > > FDT, the diffusion toolkit, when it's released later this summer) uses > > > FLIRT instead of mcflirt to realign diffusion data, the point being > > > that > > > when you have eddy currents you need affine alignment not rigid body, > > > which is more of an issue with diff data...you might want to try that > > > instead, though on this data it seems to give similar results. > > > > > > Cheers. > > > > > Steve, > > thanks for the script. I'm running on linux (suse 9.0). What I was > > doing was first using mcflirt to realign the individual runs separately > > for each diffusion direction (we generally collect 4-5 runs), taking > > the mean for each direction, then using flirt with 12 dof and mutual > > information to align across the different diffusion directions. Does > > that sound reasonable? The data that I sent were all for a single > > direction, which is presumably why the results of flirt and mcflirt are > > similar. > > > > cheers > > russ > > --- > > Russell A. Poldrack, Ph.D. > > Assistant Professor of Psychology, UCLA > > Franz Hall, Box 951563 > > Los Angeles, CA 90095-1563 > > email: [log in to unmask] > > phone: 310.794.1224 > > fax: 310.206.5895 > > web: http://www.poldracklab.org/ > > > > Stephen M. Smith DPhil > Associate Director, FMRIB and Analysis Research Coordinator > > Oxford University Centre for Functional MRI of the Brain > John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK > +44 (0) 1865 222726 (fax 222717) > > [log in to unmask] http://www.fmrib.ox.ac.uk/~steve >